JP2009053342A - Minutes preparation apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a minutes preparation apparatus capable of saving a data amount required for minutes, while improving processing efficiency when the minutes is prepared. <P>SOLUTION: The minuses preparation apparatus includes: an attendant indication section 11 for indicating information of an object attendant whose speech is recorded in attendants of an electronic meeting, and information of attendants whose speech is not recorded; a speech data receiving section 12 for receiving speech data when the attendant speaks; a speech data detection section 13 for detecting the speech data corresponding to the object attendant, which is indicated by the attendant indication section 11, in the speech data received by the speech data receiving section 12; a speech data storage section 14 for storing the speech data detected by the speech data detection section 13; a speech recognition section 15 for converting the speech data stored by the speech data storage section 14 to a text data; and a minutes preparation section 16 for preparing minutes based on the text data which is converted by the speech recognition section 15. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a minutes creation device for creating minutes.

  Conventionally, minutes of meeting minutes were recorded by the meeting creator, who wrote down the contents of the meeting participants, and typed them using a clean copy or word processor based on the notes after the meeting. It was created in the form. In making minutes based on such memos, participants usually speak freely, so it takes a lot of labor to take notes of the contents of the utterances, and the minutes maker cannot participate in the meeting. There is a fear. When the minutes maker records all the contents of the meeting, the minutes writer will be burdened even if the minutes are created while listening to the recorded contents after the meeting. It was.

  Therefore, a technique is shown in which a computer recognizes the content of a participant's speech using voice recognition technology, converts the recognized content of the speech into text, and creates an electronic minutes (for example, Patent Document 1). 2).

  In addition, as a conventional electronic conference system, a client device for each participant who participates in an electronic conference and a conference server device that manages the electronic conference are connected via a network, and the participant uses the client device. It is known that the data of a speech during its own meeting is transmitted to the conference server device, and the conference server device records the speech data while proceeding with the proceedings (for example, see Patent Document 3). Note that the utterance data is audio data transmitted / received via a network.

The conference server device is configured to create the minutes based on the speech history by converting the recorded speech data of the participant into text data using a voice recognition technology. Here, when a plurality of participants speak at the same time, if the plurality of comments are recorded as one audio data, it becomes difficult to convert the data into text data that can be correctly recognized by a person. It is common for the device to record the content of a statement for each participant.
Japanese Patent Laid-Open No. 02-206825 Japanese Patent Application Laid-Open No. 07-146919 JP 2006-050500 A

  However, although the minutes creation device such as the conference server device described above records the content of the remarks for each participant, there are participants who make only remarks that are not related to the agenda during the conference and participants who are not particularly interested. Therefore, if a conventional minutes creation device creates a minutes by recording even the contents of such participants' remarks, the amount of data required for the minutes becomes enormous. Therefore, there has been a problem that the processing efficiency when creating the minutes is reduced.

  Therefore, an object of the present invention is to provide a minutes creation apparatus that can save the amount of data required for minutes and can improve the processing efficiency when the minutes are created.

The minutes creation apparatus of the present invention includes a participant designating unit for designating information on a target participant whose speech is recorded among participants of an electronic conference and information on a participant whose speech is not recorded, and the participant speaks Audio data receiving unit for receiving audio data when the audio data is received, and audio data for detecting audio data corresponding to the target participant specified by the participant specifying unit among the audio data received by the audio data receiving unit A detection unit; a voice data storage unit that stores voice data detected by the voice data detection unit; a voice recognition unit that converts voice data stored by the voice data storage unit into text data; and the voice recognition unit And a minutes creation unit for creating minutes based on the text data converted in step (b).
With this configuration, the information of the target participant to which the remark is recorded among each participant of the electronic conference is designated, and the voice data corresponding to the designated target participant among the received voice data is saved and saved. By creating the minutes by recognizing the voice data, it is not necessary to save the voice data corresponding to the participants who are not recorded, so the amount of data required for the minutes can be saved and the minutes can be saved. Can improve the processing efficiency when creating.

Also, the minutes creation device of the present invention, when the participant designation unit designates the information of the target participant, the registered voice data corresponding to the target participant is designated, the voice recognition unit, The voice data is converted into text data in accordance with the registered voice data of the target participant designated by the participant designation unit.
With this configuration, since the voice data is converted into text data in accordance with the registered voice data of the target participant, the accuracy of voice recognition can be improved.

Also, the minutes creation device of the present invention allows the participant designation unit to designate a maximum storage capacity for saving the target participant's voice data when the target participant's information is designated, and the voice The data storage unit stores the audio data detected by the audio data detection unit so as to be within the maximum storage capacity specified by the participant specifying unit.
With this configuration, since the detected voice data is stored so as to be within the maximum storage capacity corresponding to the detected voice data, the amount of data necessary for the minutes can be used effectively.

In the minutes creation device of the present invention, the participant designating unit designates the volume of audio data corresponding to the target participant when the information on the target participant is designated, and the audio data detecting unit The sound data detected by the sound data detecting unit is adjusted so as to be the sound volume specified by the participant specifying unit.
With this configuration, by adjusting the volume of the detected audio data so as to become the designated volume, appropriate audio data is saved, so that the minutes can be created accurately.

In addition, the minutes generating device of the present invention includes an audio data reproducing unit that reproduces audio data, and the audio data detecting unit includes the audio data received by the audio data receiving unit at the participant specifying unit. The audio data corresponding to the designated target participant is detected, and the audio data received by the audio data receiving unit is reproduced by the audio data reproducing unit.
With this configuration, since the audio data is reproduced while detecting the audio data corresponding to the designated target participant, the minutes can be created and the user can listen to the contents of the agenda.

  As described above, the present invention provides a minutes creation device that can save the amount of data required for minutes and can improve the processing efficiency when creating minutes.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing a conference system according to an embodiment of the present invention. In the conference system shown in FIG. 1, four conference terminals 10 are connected to each other via a network. Henceforth, when distinguishing each of the conference terminals 10, it describes as conference terminal 10A-conference terminal 10D, and when not distinguishing, it describes as conference terminal 10.

  When conducting an electronic conference, if there are participants from participant A to participant D, each of participant A to participant D uses each of conference terminal 10A to conference terminal 10D. . Note that the conference system according to the embodiment of the present invention does not particularly have a client and a server for managing a conventional conference.

  The conference terminal 10 is a personal computer or the like. When conducting an electronic conference, the conference terminals 10 connected to the network can transmit and receive data as shown by the broken lines. For example, during the electronic conference, the conference terminal 10 transmits voice data representing the remarks of the participants of the electronic conference to all the participating conference terminals 10 excluding its own conference terminal 10 via the network. The voice data of the speech received from the conference terminal 10 is reproduced.

  In the embodiment of the present invention, transmission / reception of data used in the conference system shown in FIG. 1 is performed in accordance with IP (Internet Protocol), and therefore, an IP address is registered in the conference terminal 10.

  In addition, the conference terminal 10 creates a minutes by converting voice data of a speech received from another conference terminal 10 into text data using voice recognition technology. In addition, since the minutes creation apparatus of this invention is the conference terminal 10 if it illustrates, the meeting terminal 10 is demonstrated in embodiment of this invention.

  FIG. 2 is a block diagram of the conference terminal according to the embodiment of the present invention. The conference terminal 10 has a general computer configuration, and more specifically, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), a hard disk, A network interface 21 for connecting to the network, a speaker 22 for outputting the voice of the participant of the electronic conference, a microphone 23 for inputting the voice of the participant, and an input device 24 such as a keyboard and a mouse for inputting information from the participant And a display 25 for displaying information.

  In order to prevent ambient noise from being mixed during a call, the microphone 23 and the speaker 22 may be replaced with an earphone microphone in which a microphone and an earphone are integrated.

  As shown in FIG. 2, the conference terminal 10 includes a participant designation unit 11, a voice data receiving unit 12, a voice data detection unit 13, a voice data storage unit 14, a voice recognition unit 15, a minutes creation unit 16, a voice A data reproduction unit 17, an audio conversion unit 18, and an audio data transmission unit 19 are provided. For example, these functional units may be implemented by a module of a program executed by the CPU.

  The participant specifying unit 11 causes the user of the conference terminal 10 to specify information of at least one target participant necessary for recording the minutes of the electronic conference among the participants of the electronic conference. It has become. A target participant is a participant in which a statement is recorded among each participant of an electronic conference. The timing for designating the information on the target participant may be before the start of the electronic conference or during the electronic conference.

  FIG. 3 is an image diagram of a screen that allows a user to specify participant information. FIG. 3A illustrates a designation window 30 that allows the user to specify participant information. The participant designation unit 11 displays a designation window 30 on the display 25. The designation window 30 in FIG. 3A is for causing the information of up to eight participants to be designated via the input device 24 when, for example, a maximum of eight participants participate in the electronic conference.

  The designation window 30 has a designation item 31 for designating participant information for each participant. FIG. 3B is an enlarged view of only the designated item 31.

  The items specified by the user include a participant name item 33 representing the name of the participant, a registered voice data identification item 34 required at the time of voice recognition, an identification information item 35 for identifying a device used by the participant, There is a save item 36 for designating whether or not to save the participant's voice data. The participant name item 33, the registered voice data identification item 34, the identification information item 35, and the save right item 36 are designated by the user.

  When the conference terminal 10 conducts an electronic conference using a network compliant with IP, the identification information item 35 is the IP address of the conference terminal 10. The participant designating unit 11 registers the information of the designated identification information item 35 in the identification information of the control information shown in FIG.

  The control information shown in FIG. 4 is based on designated participant information, and is stored in the RAM or hard disk. Note that FIG. 4A illustrates an example of control information included in the conference terminal 10D. In the embodiment of the present invention, an IP address is registered for the identification information.

  As for the participants in the control information of FIG. 4A, those specified in the participant name item 33 are registered as shown in FIG. 3B, and the registered voice data of the control information in FIG. As shown in FIG. 3B, the one designated by the registered voice data identification item 34 for identifying whether the voice is for male or female is registered. In addition, as shown in FIG. 3B, what is specified in the save right item 36 is registered as to whether or not to save.

  The control information in FIG. 4B includes the presence / absence of speaker output and the maximum storage capacity in addition to FIG. 4A. The presence / absence of speaker output and the maximum storage capacity will be described later. A speaker output presence / absence item for specifying the presence / absence of speaker output and a maximum storage capacity item for specifying the maximum storage capacity are newly added to the specification item 31 (not shown). No), the items specified in the speaker output presence / absence item and the maximum storage capacity item are respectively registered in the presence / absence of speaker output and the maximum storage capacity in the control information of FIG.

  The voice data receiving unit 12 receives voice data when a participant speaks via the network interface 21 and outputs the received voice data to the voice data detecting unit 13. For example, when 8 people participate in an electronic conference, the audio data receiving unit 12 receives audio data of 7 people excluding the audio data of the participants of its own conference terminal 10.

  When an electronic conference is performed using an IP-compliant network, the voice data is set in the payload portion of the IP packet and is real-time according to a technology compliant with RTP (Real-time Transport Protocol) or the like. Received. The voice data receiving unit 12 outputs the voice data received in real time to the voice data detecting unit 13.

  The audio data detection unit 13 outputs the audio data received by the audio data reception unit 12 to the audio data reproduction unit 17. The audio data reproduction unit 17 reproduces the audio data by converting the audio data output by the audio data detection unit 13 from digital to analog, and outputs the reproduced data to the speaker 22.

  Participant's voice data excluding the voice data of the participant of his / her conference terminal 10 among the voices of the participant's speech is heard through the speaker 22.

  The voice data detection unit 13 detects voice data corresponding to the target participant designated by the participant designation unit 11 from the voice data received by the voice data reception unit 12.

  For example, when an electronic conference is performed using a network compliant with IP, since the IP address and the like are output from the participant specifying unit 11 to the audio data detecting unit 13, the audio data detecting unit 13 is shown in FIG. It is determined whether there is an IP address set in the identification information of the control information that matches the IP address in the header portion of the IP packet containing the voice data. The voice data detection unit 13 determines whether the saving of the control information shown in FIG. 4 corresponding to the IP address at the time of matching is “save”, and the participation is “save” The voice data storage unit 14 stores the voice data of the target participant who is the user.

  Further, the voice data detection unit 13 determines whether or not the saving of the control information shown in FIG. 4 corresponding to the voice data output by the voice conversion unit 18 is “save”. The voice data of the target participant who is the “

  In addition, the audio data detection unit 13 outputs the audio data received by the audio data reception unit 12 to the audio data reproduction unit 17, but the speaker output is performed according to the presence or absence of the speaker output of the control information in FIG. Only the audio data that is present is output to the audio data reproducing unit 17. When it is not desired to listen to the speech of a specific participant, unnecessary audio data is reproduced by stopping the reproduction of audio data by registering the presence or absence of speaker output of the control information in FIG. 4B. Processing efficiency can be improved when the minutes are created by omitting the processing.

  By the way, the volume meter 32 in the designated item 31 shown in FIG. 3B represents the volume of the voice data of the target participant. For example, when adjusting the volume before the start of the electronic conference or the like. Referenced.

  As a method of using the volume meter 32, when the user selects the designated item 31 that the saved item 36 is to be saved and presses the start button 41, the user speaks from the conference terminal 10 of the selected designated item 31. Are displayed on the volume meter 32. When the start button 41 is pressed, for example, the voice data detection unit 13 detects the volume level of the target participant's voice data, and displays the detected volume level on the volume meter 32 of the corresponding designation item 31. Let

  While the volume level is displayed on the volume meter 32, the participant specifying unit 11 causes the user to specify the volume with the volume adjustment button 43, so that the audio data detecting unit 13 selects the corresponding audio data. The volume is adjusted so as to be the volume specified by the participant specifying unit 11. If the user wants to stop displaying the volume level, the user presses the stop button 42.

  The clear button 44 clears the contents set in the various items of the selected designation item 31, and the save button 45 changes the contents set in each designation item 31 to the control information shown in FIG. To register.

  Here, FIG. 5 illustrates a diagram representing the flow of audio data according to the speech of each participant. The participant of the conference terminal 10A speaks from time t1, and then stops speaking and is in a silent state, and represents a state of speaking from time t4. The participant of the conference terminal 10B speaks from time t3, then stops speaking and is silent, and represents a state of speaking from time t6. The participant of the conference terminal 10D speaks from the time t2, and then stops speaking and has a silent state, and represents a state of speaking from the time t5. Note that the speech of the participant of the conference terminal 10C is omitted because it is set not to be saved in the control information of FIG.

  The audio data storage unit 14 stores the audio data detected by the audio data detection unit 13. For example, as shown in FIG. 5, when the voice data detection unit 13 detects voice data based on the speech of each target participant, the voice data storage unit 14 detects a silent part of the speech of each target participant. The voice data for the period when the speech was made is stored as an electronic file.

  As a method for detecting a silent portion, for example, there is a method in which it is regarded as silent when a certain volume level is not reached within a predetermined time.

  For example, as shown in FIG. 6, when the voice data storage unit 14 sets the speech start time to the voice data and detects a silent part of the speech after the speech is started, the speech data storage unit 14 is silent after the speech is started. Audio data up to the state is saved as an electronic file so that the target participant can be identified.

  The audio data when the target participant of the conference terminal 10A speaks from time t1 is referred to as audio data A1. The audio data A1 is stored in a recording medium such as a hard disk by the audio data storage unit 14 as an electronic file A1. Also, the audio data when the target participant of the conference terminal 10A speaks from time t4 is set as audio data A2. The audio data A2 is stored in a recording medium such as a hard disk by the audio data storage unit 14 as an electronic file A2.

  The audio data when the target participant of the conference terminal 10B speaks from time t3 is referred to as audio data B1. The audio data B1 is stored in a recording medium such as a hard disk by the audio data storage unit 14 as an electronic file B1. Also, the audio data when the target participant of the conference terminal 10B speaks from time t6 is referred to as audio data B2. The audio data B2 is stored in a recording medium such as a hard disk by the audio data storage unit 14 as an electronic file B2.

  The audio data when the target participant of the conference terminal 10D speaks from time t2 is set as audio data D1. The audio data D1 is stored in a recording medium such as a hard disk by the audio data storage unit 14 as an electronic file D1. Also, the audio data when the target participant of the conference terminal 10D speaks from time t5 is referred to as audio data D2. The audio data D2 is stored in a recording medium such as a hard disk by the audio data storage unit 14 as an electronic file D2.

  Note that the file name of the electronic file provided by the audio data storage unit 14 may be a combination of the start time and information for identifying the participant.

  The voice recognition unit 15 recognizes the voice data stored by the voice data storage unit 14 and converts it into text data. As shown in FIG. 7, the speech recognition unit 15 has a speech recognition engine having a learning function, and recognizes speech recognition accuracy (voice recognition) by recognizing speech data of the same target participant as much as possible. Rate).

  For example, as shown in FIG. 7, the voice recognition unit 15 recognizes only the voice data of the target participant of the conference terminal 10A within a predetermined time, and the voice of the target participant of the conference terminal 10B within the next fixed time. Only the data is voice-recognized, and only the voice data of the target participant of the conference terminal 10D is voice-recognized within the next fixed time, and the voice-recognized result is converted into text data.

  Further, the voice recognition unit 15 has registered voice data indicated by the registered voice data identification item 34 set in the control information of FIG. 4B, and the target participant specified by the participant specifying unit 11 The speech recognition is performed based on the registered speech data, and the speech recognition result is converted into text data, thereby further improving the speech recognition accuracy (speech recognition rate).

  In FIG. 4, for the registered voice data identification item 34, information for identifying whether the voice is for male or female is specified in order to improve the efficiency of voice recognition. It is desirable to register the registered voice data obtained by recording the participant's voice in advance and set the registered voice data in the registered voice data identification item 34. If it does in this way, the voice recognition part 15 will improve a recognition rate by carrying out voice recognition based on registration voice data for every participant.

  The minutes creation unit 16 creates minutes based on the text data converted from the voice data by the voice recognition unit 15. As shown in FIG. 6, since the start time is set for the sound data stored by the sound data storage unit 14, the minutes are created by rearranging the text data converted from the sound data in the order of the start time. It is supposed to be.

  FIG. 8 shows a diagram when the minutes created by the minutes creation unit 16 are displayed on the display 25. FIG. 8A shows the minutes corresponding to the HTML (Hyper Text Markup Language) format, and FIG. 8B shows the minutes corresponding to the text format.

  The minutes creation unit 16 may create the minutes in real time during the electronic conference, or may create the minutes at an arbitrary time such as when the electronic conference is over.

  In FIG. 8A, the speaker and the message content are displayed in correspondence with each other. Further, in FIG. 8A, the speech data of the utterance content is reproduced by pressing a playback button corresponding to the utterance content. For example, when a playback button is pressed, the audio data playback unit 17 acquires an electronic file of audio data corresponding to the pressed playback button from the audio data storage unit 14 and plays it back.

  In FIG. 8B, the speaker and the speech start time are displayed in parentheses, and the speaker and the content of the speech are displayed in correspondence with each other. Note that the participant name designated in the participant name item 33 in FIG. 3B is displayed as the speaker in FIGS. 8A and 8B.

  The voice conversion unit 18 performs analog-to-digital conversion on the voice signal when the microphone 23 picks up the speech of the participant, and outputs the converted digital voice data to the voice data transmission unit 19 and the voice data detection unit 13. It is like that.

  When the audio data is output by the audio conversion unit 18, the audio data transmission unit 19 transmits the audio data to all the conference terminals 10 participating in the electronic conference excluding its own conference terminal 10 via the network interface 21. It is like that.

  An example of the operation of the conference terminal 10 configured as described above will be described below with reference to FIG. FIG. 9A is a flowchart showing a flow of an operation for registering information of participants in an electronic conference at a stage before starting the electronic conference.

  First, before starting an electronic conference, a user who uses the conference terminal 10 designates a participant of the electronic conference, and information on the participant is registered in the conference terminal 10. For example, as shown in FIG. 3, the participant designation unit 11 displays a designation window 30 on the display 25, and the user inputs information on various items in the designation item 31 via the designation window 30 using the input device 24. input. Information on the designated item is registered in the control information of FIG. 4 by the participant designation unit 11 (step S1).

  FIG. 9B is a flowchart showing an operation flow when audio data is received after the electronic conference is started.

  After the electronic conference is started, the voice data receiving unit 12 outputs the received voice data to the voice data detecting unit 13 when received via the network interface 21 (step S11). The voice data detection unit 13 acquires an IP address from the header part of the IP packet that includes the output voice data (step S12).

  The voice data detection unit 13 determines whether or not to save the voice data corresponding to the identification information when the same IP address acquired in step S12 is in the identification information of the control information shown in FIG. S13). If the identification information of the control information shown in FIG. 4 matches the IP address acquired in step S12, and the saving of the audio data corresponding to the identification information is “save”, the voice The data detection unit 13 outputs the audio data to the audio data storage unit 14.

  When the voice data storage unit 14 detects a silent part of the speech of each target participant, the speech data storage unit 14 separates the speech data of the speech period (step S14), and information on the start time of the speech for each separated speech data Is set (step S15). Next, the audio data storage unit 14 stores the audio data in which the time information is set in a recording medium (step S16).

  When a value representing the maximum storage capacity is set in the maximum storage capacity of the control information shown in FIG. 4B, the audio data storage unit 14 until the set maximum storage capacity is reached. Save the audio data of the target participant. When the volume of the stored audio data reaches the maximum storage capacity of the control information corresponding to the target participant, the audio data storage unit 14 stops storing the audio data. When “no restriction” is set as the maximum storage capacity of the control information, the audio data storage unit 14 stores the audio data of the target participant without limitation.

  On the other hand, the audio data received by the audio data receiving unit 12 is reproduced by the audio data reproducing unit 17 regardless of whether the audio data is stored in step S13 (step S17). Output as audio.

  The presence / absence of the speaker output of the control information shown in FIG. 4B is information indicating whether or not the voice uttered by the participant is output to the speaker 22, and the presence / absence of the speaker output of the control information is “none”. The audio data of the participant set to “” is not reproduced by the audio data reproduction unit 17.

  FIG. 9C is a flowchart showing an operation flow when audio data is stored. The flowchart shown in FIG. 9C shows the operation when creating the minutes in real time.

  When the voice data is stored in the recording medium by the voice data storage unit 14 (step S21), the voice recognition unit 15 displays the registered voice indicated by the registered voice data identification item 34 set in the control information of FIG. Based on the data, the speech data stored by the speech data storage unit 14 is recognized as speech and converted to text data (step S22).

  Next, the minutes creation unit 16 creates the minutes by rearranging the text data converted from the voice data in order from the oldest to the start time (step S23), and the created minutes are shown in FIG. As shown, it is displayed on the display 25 (step S24).

  As described above, the conference terminal 10 causes the information of the target participant to be recorded to be recorded among the participants of the electronic conference, and the audio corresponding to the specified target participant in the received audio data. Since only the data is saved and the saved voice data is recognized and the minutes are created, it is no longer necessary to save the voice data corresponding to the participant whose speech is not recorded. Can be saved.

  In addition, the conference terminal 10 creates the minutes by creating the minutes only from the speech data of the necessary participant's utterance, so that it is not necessary to recognize the voice data of the unnecessary participant's utterance. When processing efficiency can be increased.

  In addition, since the registered voice data of the target participant is registered in the conference terminal 10 and the voice recognition unit 15 recognizes the voice according to the registered voice data registered and converts the voice data into text data, the conference terminal 10 The accuracy of voice recognition can be improved. Since only the registered voice data of the target participant needs to be registered, it is not necessary to register unnecessary registered voice data that does not record a speech.

  In addition, since the audio data storage unit 14 stores the audio data of the target participant until the set maximum storage capacity is reached, and does not store exceeding the maximum storage capacity, the conference terminal 10 stores data necessary for the minutes. The amount can be used effectively.

  In addition, since the audio data detection unit 13 adjusts the volume of the corresponding audio data so as to be the volume specified by the participant specifying unit 11, appropriate audio data is stored, so that the conference The terminal 10 can accurately create the minutes.

  In the embodiment of the present invention, all of the conference terminals 10A to 10D may take minutes, one or more conference terminals 10 may take, and the conference terminal takes minutes. The number of 10 is not limited. In addition, since the conference terminal according to the embodiment of the present invention creates minutes only by the speech of the participants that are necessary for each conference terminal 10, various minutes are created for each conference terminal 10 even in the same conference. be able to. Further, the terminal for collecting the minutes may be the conference terminal 10D, and the conference terminals 10A to 10C may be replaced with a general personal computer. That is, one of the conference terminals is the conference terminal 10 and the remaining terminals. The configuration may be such that the computer is a general personal computer.

  Further, as shown in FIG. 10, the conference terminal 10 may be connected to the USB hub 40, and each earphone microphone 50 may be connected to the USB hub 40. In the embodiment of the present invention, the identification information is described as an IP address. However, the identification information when implemented in the form shown in FIG. 10 is a device identifier of the earphone microphone 50.

  In addition, during the electronic conference, the user deletes the information in the specified item 31 via the specified window 30 so that the participant specifying unit 11 cancels the participant registered in the control information. You can also.

  Also, during the electronic conference, the user changes the right or wrong of saving in the specified item 31 via the specified window 30 from “Save” to “Do not save”, so that the participant specifying unit 11 can use It is also possible to register “do not save” whether or not the save has been changed by the user. Conversely, during the electronic conference, the user changes the right or wrong of the saving in the designated item 31 via the designation window 30 from “Do not save” to “Save”, so that the participant designation unit 11 It is also possible to register “Save” whether or not the save operation has been changed by the user.

  As described above, the present invention can save the amount of data required for the minutes, has the effect of improving the processing efficiency when creating the minutes, and can be applied to a personal computer for electronic meetings. Useful.

The figure showing the conference system which concerns on embodiment of this invention Block diagram of a conference terminal according to an embodiment of the present invention Image of screen that lets users specify participant information The figure which shows the control information based on the information of the designated participant A diagram showing the flow of audio data for each participant Diagram of saving audio data as an electronic file from when speech is started until silence is reached when a silent part of speech is detected after speech is started Image of recognizing voice data of the same participant as much as possible The figure which shows the minutes which rearranged the text data which is converted from voice data in order of the start time The flowchart which shows an example of operation | movement of the conference terminal which concerns on embodiment of this invention. The figure showing the conference system which concerns on a different form from embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Conference terminal 11 Participant designation | designated part 12 Audio | voice data receiving part 13 Audio | voice data detection part 14 Audio | voice data preservation | save part 15 Audio | voice recognition part 16 Minutes preparation part 17 Audio | voice data reproduction part 18 Audio | voice conversion part 19 Audio | voice data transmission part 21 Network interface 22 Speaker 23 Microphone 24 Input device 25 Display 30 Specified window 31 Specified item 32 Volume meter 33 Participant name item 34 Registered voice data identification item 35 Identification information item 36 Save item 40 USB hub 41 Start button 42 Stop button 43 Volume adjustment button 44 Clear button 45 Save button 50 Earphone microphone

Claims (5)

Participant designating section for designating information of target participants whose utterances are recorded and information of participants whose utterances are not recorded among the participants of the electronic conference,
An audio data receiving unit for receiving audio data when the participant speaks;
Among the audio data received by the audio data receiver, an audio data detector that detects audio data corresponding to the target participant specified by the participant specifying unit;
An audio data storage unit that stores the audio data detected by the audio data detection unit;
A voice recognition unit for converting the voice data stored by the voice data storage unit into text data;
A minutes creation device, comprising: a minutes creation section for creating minutes based on the text data converted by the voice recognition section. When the participant designation unit designates the information of the target participant, the registered voice data corresponding to the target participant is designated,
2. The minutes creation apparatus according to claim 1, wherein the voice recognition unit converts voice data into text data in accordance with registered voice data of a target participant designated by the participant designation unit. The participant designation unit designates the maximum storage capacity when saving the target participant's voice data when the target participant's information is designated,
The audio data storage unit stores the audio data detected by the audio data detection unit so as to be within a maximum storage capacity specified by the participant specifying unit. 2. Minutes preparation device described in 2. When the participant specifying unit specifies the information of the target participant, the volume of audio data corresponding to the target participant is specified,
The sound data detection unit adjusts the sound volume of the sound data detected by the sound data detection unit so as to be a sound volume specified by the participant specifying unit. Minutes creation device described in any of the above. Provided with an audio data reproduction unit for reproducing audio data,
The voice data detection unit detects voice data corresponding to a target participant designated by the participant designation unit from the voice data received by the voice data reception unit, and receives the voice data by the voice data reception unit. The apparatus for creating minutes as claimed in claim 1, wherein the audio data reproducing unit reproduces the audio data thus reproduced.

Images